Moving to ServiceNow License Optimization: Migration Considerations
ServiceNow Software Asset Management (SAM) tracks software licenses, compliance and optimization. This page sets out what to inventory, map, cut over and reconcile when a team moves its software license records into ServiceNow. It covers the data classes that move at cutover, the exit criteria for a reconciliation gate, and the signs that a migration should wait.
What to Inventory Before Migration
Document your current software licensing situation: which software products does your organization use, how many licenses of each do you own, what are the terms (perpetual, maintenance contracts, subscriptions)? Document how you currently track licenses: manually in spreadsheets, in a separate tool like Flexera? Inventory any automations you currently use to discover deployed software. Assess your current compliance status: have you been audited recently, are there known gaps or concerns? Interview stakeholders: procurement teams, finance teams, IT operations and department heads to understand their needs and pain points with current license management.
Data and Process Mapping for Migration
Map current entitlements (products, quantities, terms, costs) to ServiceNow Software Asset Management (SAM) data structures. Determine which data fields are critical and which are optional. Map current deployment tracking (if any) to Software Asset Management discovery and asset management. If you have existing audit trails or compliance documentation, plan how to migrate or reference that history. Document current workflows: how are licenses requested, approved and provisioned? How are license renewals handled? How are compliance gaps addressed? These workflows may need to change in Software Asset Management.
Cutover Options and Strategies
Big bang cutover: shut down the old system and cut over to Software Asset Management on a specific date. This is fastest but risky: if critical data is missing or Software Asset Management is not ready, you lose access to historical data. Big bang suits simpler environments or teams with strong confidence in their migration plan. Parallel run: run the old system and Software Asset Management side-by-side for a period, comparing results. This is safer but requires maintaining two systems and can be confusing for users. Often a practical choice for organizations with complex entitlements. Phased cutover: migrate one product family at a time (for example, Microsoft, Adobe, then specialized tools). This reduces risk and allows time to refine the process but prolongs the transition.
Risks, Mitigation and Validation
Risk: loss of historical entitlement or compliance data. Mitigation: export all current data and archive it separately before migration. Risk: entitlement data is incomplete or inaccurate. Mitigation: audit current entitlements before migration. Clean up data before importing to Software Asset Management. Risk: software discovery in Software Asset Management differs from current process, leading to different deployment counts. Mitigation: run both discovery methods in parallel during pilot and compare results. Review the discovery models and their normalization in Software Asset Management. Before cutover: validate that all entitlements are imported into Software Asset Management with correct data. Validate that software discovery captures expected deployments. Validate that compliance calculations match manual calculations.
Cutover Planning and Data Reconciliation
Plan the cutover date early. Coordinate with finance and procurement teams to choose a date outside peak spending periods. A cutover during fiscal year-end or budget season risks disruption. A cutover after peak season is usually safer. For each type of license, identify the data migration path: perpetual licenses convert to terms if that is the new model; support contracts may be consolidated or simplified. Create a detailed cutover checklist: what data moves from the old system to ServiceNow? Which reports must be recreated? Which integrations must be reconfigured?
Vendor Transition and License Consolidation
During migration, review vendor relationships. Record license terms from each vendor as software entitlements in ServiceNow where possible. Renegotiate volume discounts: consolidating purchase agreements often results in lower costs. Identify vendors who provide duplicate functionality; eliminating duplicate vendors reduces complexity. Document all license agreements: who provides the license? What are the terms (perpetual, term-based, user-based)? What are the support and renewal dates? This documentation ensures nothing is lost during the transition.
During the migration, run parallel testing: process licenses through both the old system and ServiceNow to verify that ServiceNow produces the same results. This validates that data mapping is correct before cutover.
Maintain vendor contracts and renewal schedules. Set calendar reminders for contract anniversaries so renewal negotiations can start early. Early negotiation gives more room for better terms than waiting until renewal is imminent.
Organizational Readiness and Change Management
Migrating to Software Asset Management is a significant change. Procurement teams must learn new processes for tracking and managing software licenses. Finance teams must adjust their budgeting and reporting. IT teams must update their inventory processes. Plan change management: communicate what is changing and why. Conduct training for key users. Identify champions who can help others understand and adapt to the new system. Change management success is often the difference between a migration that succeeds and one that fails.
Post-Migration Optimization and Continuous Improvement
After the initial migration, continue to refine the system. Use the early period after go-live to identify gaps and make adjustments. Engage with teams using the system to gather feedback. Implement improvements based on feedback. After a fixed review point that you choose in advance, conduct a full review: are we achieving our goals? Is license utilization improving? Are compliance risks reducing? Use the review results to adjust strategy going forward.
Deciding what moves now and what stays put
Migration plans often treat the license estate as one block of data. It is not one block. Some records change often, such as installations and user assignments. Others change once a year, such as perpetual entitlements and maintenance terms. A third group is historical and rarely edited at all, such as closed agreements and past audit outcomes. Each class carries a different risk on the day of the switch, so each needs its own answer to the question of whether it moves now.
| Data class | Move at cutover? | Reason to decide it this way |
|---|---|---|
| Active agreements with live renewal dates | Yes, after reconciliation | Renewal dates must be held in the new system before the old one is switched off |
| Installations and assignments | Yes, from a fresh discovery run | Old snapshots can go stale quickly, so a new baseline is often more reliable than a copy |
| Closed or expired agreements | Archive, do not load as active | Keeping them active inflates the entitlement total and invites false matches |
| Past audit outcomes and correspondence | Keep in an accessible archive | These answer questions that may be asked long after the cutover date |
The table is a starting point. Your own contract mix may move a class up or down the list, and the decision should be recorded so a future reviewer can trace why a record was archived instead of loaded.
A reconciliation gate with written exit criteria
The point at which a migration is ready is not a calendar date. It is the point at which the new data matches the old on terms that were agreed in advance. Write the exit criteria before the load begins, so that nobody is tempted to lower the bar once the work is under way.
- Agree the match fields for each class, such as agreement reference, product, metric, quantity and term end.
- Load a test set and reconcile it field by field against the old register, recording every difference and its cause.
- Classify each difference as a data error to correct, a mapping difference to document, or a genuine change in meaning to escalate.
- Repeat the load on a fresh extract, since a single test set can hide problems that only appear in a full population.
- Obtain sign-off from the finance and procurement owners of the figures, not only from the team that performed the load.
- Keep the old system readable for a defined period after cutover so that any later question can be answered from the source.
Keeping evidence continuous across the switch
The compliance story should not break at the cutover line. Before the switch, export the old audit history together with the record of who changed what and when, and store it where it can be retrieved without the old platform running. A new platform that starts with an empty history makes its first report look more trustworthy than the evidence behind it warrants, and the hole it leaves is one an auditor is likely to find. The migration reconciliation controls guide covers how source and target records are compared before closure.
When a reconciled figure still does not match
Some differences survive the corrections. Often the two systems hold the same fact in different forms, for example a quantity recorded as seats in one and as installed instances in the other. Do not force the match. Document the mapping, show the calculation both ways, and get the owner to choose the basis that the contract actually uses. A figure that matches only because a field was changed to fit is usually worse than a documented difference, because the difference is likely to reappear at the next audit with no explanation attached.
Signs the migration should wait
- A major renewal or true-up is due inside the cutover window, and the team has not agreed which system owns the negotiation figures.
- Discovery in the new platform has not yet been validated against a known sample of installations.
- The finance owner is not available to sign off the reconciliation within the window the plan allows.
- An integration that feeds procurement is still in build, so renewals entered after cutover would not reach the new register.
The integration point deserves its own plan. The integration patterns guide covers how usage and purchase data flow into the module, and the integration cutover readiness article covers what readiness means for an interface before a controlled cutover.
Should the old register be deleted once the new one is live?
No. Hold it in a readable state for an agreed window, then move it to archive under the retention rule for license and audit records. Deleting it removes the source that any later reconciliation depends on.
Can a Software Asset Management Professional move be run as the same project?
It can share governance and data work, but the reconciliation gate should be run separately for each product family, so that a problem in one does not hold up the others. The Software Asset Management Professional migration considerations guide covers what to inventory from the current system before a move.
Who should sign off the cutover?
The owners of the figures that will be reported, with the migration lead in support. Sign-off by the team that performed the load alone is a common reason a reconciled dataset is later distrusted.
Reworking a date after it has been announced usually costs more than settling the gate early.
Software license migration: tables, fields and values to map
A software license migration moves entitlement, contract and installation records into ServiceNow so that licenses can be compared with what is deployed. The deciding fact is the record trail: each entitlement should link to a product, a publisher and, where one exists, a contract before any deployed software is counted against it.
These are common out-of-box names, so check them on your own instance and release, because labels and defaults change between releases. Some of the tables appear only when the software asset management application is installed.
| Name | What it holds |
|---|---|
alm_license |
Software entitlement records (labelled Software License before the Professional migration) |
cmdb_sam_sw_install |
Software installation records, typically populated by discovery or another data source |
ast_contract |
Contract records that a license can reference |
Example (hypothetical): a team keeps one spreadsheet row per purchase order. Before loading anything, it adds a column for the publisher company, one for the product it will map to, and one for the contract it belongs to. Rows that cannot fill all three are held back in a review list instead of being loaded half-linked, and each loaded row keeps its original spreadsheet row number in a note so later questions can be traced.
Which records should be loaded first?
Load companies and contracts first, then software models, then licenses. Each later record points at an earlier one, so loading in that order avoids orphaned references and lets you test the links on a small batch before the full load.
Why do installs not match licenses after the load?
A common cause is that discovered software was not normalized to a software model, so an install has nothing to match against. Check normalization and the publisher mapping before blaming the entitlement data.
Can the spreadsheet be kept as a fallback?
Yes, as a frozen, read-only copy with a recorded date, kept until the first reconciliation is accepted. Editing it after cutover recreates the two-registers problem the migration was meant to remove.
For the product overview, see software license optimization, the Software Asset Management Professional guide and the software asset compliance workflows.
What Software Asset Management Does and How to Decide Whether It Fits
License optimization in ServiceNow Software Asset Management means comparing what your organization is entitled to use with what is actually installed, then acting on the gap. In ServiceNow, the Software Asset Management application does this comparison inside the platform. License records are matched against discovered software installations, so the compliance position and any unused entitlements can be read from one place instead of being rebuilt in a spreadsheet each quarter.
The decision rule is simple. If you hold licenses from more than one publisher, face a renewal or true-up within the next year, and your install data comes from discovery rather than manual collection, the platform will usually repay the effort. If your estate is small, you have a handful of contracts and nobody runs regular discovery, a well-kept register is often enough. An optimization project in that case only adds a second place for data to go stale.
In most implementations the work falls into three parts. First, each entitlement is recorded with its product, publisher, quantity, metric and term end date. Second, installations are normalized so that each discovered product maps to a product model. Third, the compliance view is reviewed on a schedule by the people who own the contracts, not only by the team that runs the tool. Skipping the second step is a frequent reason the first report looks wrong.
A common mistake is to treat the installation count as the number of licenses you need. The metric in the contract decides that. A per-user metric counts people, a per-device metric counts machines, and a core-based metric counts processor capacity. Check the metric on your own contract, then confirm how your instance and release report each one, because labels and defaults can differ between releases.
Maximizing the value of a ServiceNow license investment starts on the spend side. Installs nobody uses and entitlements nobody has assigned are often where savings come from, and they show up when the entitlement and install data agree. The two levers are reclaiming unused rights and buying new rights. For a plain-language overview of the module, see the ServiceNow IT asset management guides. To see how software models link installs to a product, read how the ServiceNow software model works. If your install data comes from discovery, the ServiceNow Discovery guides explain how that data is collected.
Deciding in Advance When to Roll Back a License Cutover
A rollback plan answers one question before the load starts: at what point does the team stop moving forward and return to the old register, and who has the authority to make that call. Teams that leave this open usually end up deciding it late on cutover day, under deadline pressure, with someone asking whether to continue.
Separate two kinds of failure first. A load failure stops the process: a file is rejected, a mapping throws errors, or a required field is empty across a whole class of records. These are visible and usually fixed by correcting the source and reloading. A reconciliation failure is quieter. The load completes, but agreement counts, product matches or term end dates fall outside the tolerance agreed at the gate. This is the case that needs a rollback decision, because the new register is now showing figures that nobody has verified.
A workable rollback plan usually settles four things:
- A decision owner from finance or procurement, not only from the migration team.
- A cut-off time after which the old register remains the system of record for the current cycle.
- A record of which data has already been written to the new platform, including the import sets the records came through, so they can be found. The removal method depends on your instance and release, so test it in a non-production environment first.
- A list of every renewal alert, approval and report that was switched off on the old side and must be restored.
The last item is easy to miss. A renewal reminder is often turned off on the old side to avoid duplicate alerts. If the cutover is reversed, that control disappears without anyone deciding it should. Keep the old process running until the new one has produced at least one full cycle of outputs that were checked against it.
The reconciliation gate and the rollback plan work as a pair. The gate says when the data is good enough to switch. The rollback plan says what happens if the switch turns out to be wrong after the old system has gone quiet. A migration with only the first has no clear answer when figures drift later. The cutover approach for reconciled data shows how the two are built together, and the migration services page covers the wider sequence.
Frequently asked questions
What are common challenges?
Common challenges include inadequate planning, insufficient resources and unclear ownership.
How should we approach this?
Start with assessment, establish clear ownership and follow a phased approach.
What should we avoid?
Avoid rushing implementation, skipping testing and failing to document processes.
Related guides
Related Oblytech services
Where this topic turns into delivery work, these are the Oblytech services that cover it.
Planning, running or fixing a ServiceNow programme? Tell us where you are and we will suggest a sensible next step.
Contact us